The present invention relates to an atomic absorption spectrophotometer and in particular to an atomic absorption spectrophotometer adapted to analyze microelements dissolved in a solution sample.
In accordance with the prior art of this field, for example, as shown in U.S. Pat. No. 4,098,554, light emitted from a hollow cathode lamp serving as a light source is led into an electric furnace. In this electric furnace, the solution sample is dried, ashed and atomized, and light from the light source is radiated on the atomized vapor to cause atomic absorption. Atomic absorption lines of the light transmitted through the electric furnace without being absorbed are selected by a monochromator and processed by a detection system. Numercal values proportionate to the densities of elements to be analyzed in the sample are thus obtained.
The structure of a conventional electric furnace used in such an atomic absorption spectrophotometer is shown in FIGS. 1 and 2.
In FIGS. 1 and 2, a cylindrical heating material 11 is put between a right-hand electrode 12 and a left-hand electrode 13 and electrically connected to these electrodes. A current is supplied through a right-hand terminal 14 and a left-hand terminal 15 to heat the cylindrical heating material 11 to a needed temperature.
The right-hand electrode 12 has a window 16 comprising a quartz material, for example, and transmitting a light ray 22 used for the atomic absorption measurement. The left-hand electrode 13 also has a window 17 comprising a quartz material for the incidence of a measuring light ray 23. A cylinder 18 comprising an insulator is put on the right-hand electrode 12 and the left-hand electrode 13. Airtightness of the contacting portion is maintained by O-rings 24 and 25. The right-hand electrode 12 has a tube 21 for introducing inert gas, through which argon gas, for example, is introduced. The central part of the cylinder 18 has an opening 19 for introducing a sample to be analyzed. The cylindrical heating material 11 also has an opening 26 which is coaxial with the above described opening. A liquid drop 20 of 20 .mu.l, for example, is injected into the inside of the cylindrical heating material 11 via the above described openings 19 and 26 by using a micro-syringe, for example. Dried argon gas of a constant flow rate flows in via the tube 21 and flows out via the openings 26 and 19. Air within the cylindrical heating material 11 is thus expelled. As a result, reaction of the cylindrical heating material 11, heated to as high as 3,000.degree. C. reacts with oxygen and wasted of the material are prevented.
As the cylindrical heating material 11 shown in FIGS. 1 and 2, a small cylinder, which is 40 mm in length, 5 mm in internal diameter, and 8 mm in external diameter, comprising a graphite material of high purity has heretofore been used.
When a current of 400 amperes with AC 12 volts is supplied to this cylindrical heating material 11, its surface temperature reaches approximately 300.degree. C. If oxygen gas exists under this state, the cylindrical heating material violently reacts with the oxygen gas to produce carbonic acid gas and be wasted. In the actual analyzing operation, the relationship between the current supplied to the cylindrical heating material 11 and the temperature is derived beforehand, and a predetermined heating program is made. The cylindrical heating material 11 is then heated in accordance with the predetermined heating program.
FIG. 3 shows an example of the heating program for heating the electric furnace of FIG. 1.
The sample water drop 20 of, say, 20 .mu.l shown in FIG. 2 is heated by the heating program shown in FIG. 3 as described below.
(a) Drying stage: The temperature is kept at 90.degree. C. for 60 seconds to evaporate water used as a solvent and to perform drying.
(b) Ashing stage: The sample is heated to a temperature of 600.degree. C. for 10 seconds to ash organic matters included in the dried solute.
(c) Subsequent atomizing stage: The sample is heated to a temperature of 2,800.degree. C. for 10 seconds to evaporate the ashed inorganic ingredients. The inorganic ingredients undergo pyrolysis to be dissociated into atomic states. In FIG. 1, the incident measuring light ray 23 from the window 17 is absorbed into the inside of the cylindrical heating material 11 by atoms generated in the above described atomizing stage. The remaining light ray 22 is then transmitted to the outside of the electric furnace by window 16.
By the above described method, the concentrations of elements dissolved in the sample water drop 20, such as the cadmium element, are measured.
The prior art heretofore described has the following problems:
(1) The time required for drying is long: In the above described heating program for drying, ashing and atomizing the liquid drop 20 of 20 .mu.l injected into the cylindrical heating material 11 of the electric furnace of the prior art shown in FIG. 1, the time required for drying is 60 seconds. If the liquid drop 20 is suddenly boiled and scattered in the drying process, the scattered liquid drop sticks to internal walls of the electrodes 12 and 13 or sticks to internal faces of the windows 16 and 17. Since the temperature is comparatively low at these portions, it is impossible to dry the sample in 60 seconds. The subsequent ashing and atomizing stages cannot be substantially performed. In the drying stage, therefore, the temperature must be set to such a temperature that the liquid drop 20 will never be suddenly boiled, i.e., 90.degree. C. which is sufficiently lower than the boiling point. As a result, it takes approximately 60 seconds for the liquid drop to be dried. In the subsequent ashing stage and the atomizing stage, the temperature can be set to a high value because the sudden boiling phenomenon does not occur. Accordingly, the duration of each of the ashing stage and the atomizing stage is as short as 10 seconds.
Since the drying time is as long as 60 seconds as described above, it takes 80 seconds to analyze one sample. This results in a drawback that the analysis can be performed only 45 times at most in one hour. In reality, it takes approximately 10 seconds to inject the sample by using the micro-syringe, resulting in a further lowered work efficiency.
(2) The sensitivity of atomic absorption is low: In an atomic absorption spectrophotometer of the prior art, a well-known hollow cathode lamp is used as the light source . This lamp is filled with inert gas such as neon gas or argon gas with pressure typically ranging from 4 to 12 Torr. In the discharge state, the atomic spectrum of the cathode material is generated and emitted to the outside.
As is well known, profiles of the atomic emission line and atomic absorption line depend upon the atmosphere wherein the atoms are placed as well as the pressure and temperature of the gas. Measurement data obtained by experiments and theoretical formulas have been reported as follows.
As a report on the measurement of profiles of the atomic spectrum of a hollow cathode lamp and the atomic spectrum emitted from an acetylene flame placed in the atmospheric pressure, the following literature can be mentioned: [1] H. C. Wagenaar. and L. de Galan, Spectrochimica Acta, Vol. 28B, pp. 157-177 (1973).
As a report on the measurement of the relationship between the profile of the atomic spectrum of a light source and the atomic absorption efficiency, the following literature can be mentioned: [2] K. Yasuda, Anal. chem, Vol. 38, pp. 592-599 (1966).
As a known example of measurement of dependence of the profile upon the degree of vacuum, the following literature can be mentioned. Within a room having a degree of vacuum adjustable in the range of 50 to 760 Torr by the connection of a vacuum pump, oxygen and acetylene were burned with mixed flames. Zinc (Zn) atoms were introduced into the flames and the profile of the atomic absorption line of Zn-307.509 nm of zinc was measured. Dependence of this profile upon the degree of vacuum was thus measured. [3] T. J. Hollander and H. P. Broida, Combustion and Flame, Vol. 13, pp. 63 to 70 (1969).
The contents of these literature items can be summarized as follows: As the pressure of the atmospheric gas is raised, the breadth of the atomic spectrum is widened (as described in literature item [3]) and the peak of the spectrum is shifted into the longer wavelength direction.
In literature item [1], the atomic spectrum at a wavelength of 396.1 nm of aluminum emitted from a hollow cathode lamp is compared with that emitted from acetylene flames burning in the atmospheric pressure (760 Torr) as shown in FIG. 4. In FIG. 4, the ordinate represents the intensity of light in a relative scale. The abscissa represents the frequency of light. In FIG. 4, .alpha. represents low-pressure emission line of Al 396.1 nm emitted from a hollow cathode lamp with pressure of 8 Torr, while .beta. represents the atmospheric-pressure emission line of the same Al 396.1 nm emitted from acetylene flames with pressure of 760 Torr.
According to literature item [1], the atomic spectrum of Al 396.1 nm emitted from a hollow cathode lamp has a breadth of 86.+-.1 (.times.10.sup.-3 cm.sup.-1). The atomic spectrum of the same Al 396.1 nm emitted from acetylene flames burning in the atmospheric pressure have a breadth of 301.+-.3 (.times.10.sup.-3 cm.sup.-1). In addition, the latter is shifted by 48.+-.2 (.times.10.sup.-3 cm.sup.-1) in the longer wavelength direction (i.e., in the lower frequency direction) as compared with the former one.
The efficiency of light absorption by atoms becomes the highest when the profile of the absorption line of the atoms to be analyzed coincides with that of the emission line of the atoms used as the light source.
In the prior art apparatus shown in FIGS. 1 and 2, therefore, the emitting atoms exist in a hollow cathode lamp having a pressure of 8 Torr, and atoms to be analyzed exist in an electric furnace operating in the atmosphere of argon gas.
When aluminum (Al), for example, is analyzed by using this apparatus, the relationship between the emission line of the light source and the absorption line of the sample atoms is extremely close to that shown in FIG. 4. Absorption is caused only in the portion where the two spectra overlap each other. It is understood that the efficiency of absorption of light, i.e., the sensitivity of the atomic absorption, is largely lowered because of large discrepancy between both spectra.